With the rapidly evolving technology, there is an increase
in hacking of the secure defense databases. Nowadays, criminals use this technology
to commit crimes and to avoid being caught. So, there is a need for new and
efficient methods for preventing crimes.
Data mining is considered as a major tool for investigating,
restricting and evading crime. The data mining techniques include entity
extraction, clustering, association rule mining, decision trees, support vector
machines, naive Bayes rule, neural networks and social network analysis. These
techniques can help in accelerating the
process of crime solving and also aid computerized systems in automatic
detection of criminals.
This method is commonly used to automatically associate
people, institutions and personal details in unstructured data like police
reports. Entity Extraction provides basic information, but it can escalate the
investigation by providing minute details from large volumes of unstructured
Some of the approaches for Entity Extraction
are: machine learning, statistic-based, rule-based, and lexical lookup.
To detect whether crimes have been committed by
the same group of individuals or not, probability density function is used to
extract entities from a bunch of records. This transforms a high dimensional
vector table into an input for a police operable tool.
This helps the crime analysts and investigators
to clearly understand all undergoing investigations.
2) Cluster Analysis
techniques are used to group similar characteristics together in classes in
order to gain intelligence by maximizing or minimizing similarities.”
They can be
used to spot criminals committing crimes in similar ways. Clustering techniques
are applied through conceptual space algorithms to figure out criminal dealings
by cross referencing entities in criminal records.
Cluster Analysis along with Geographic
Information System (GIS) can be used to detect crime hot spots.
The Coplink system is considered as one of the most
successful implementation of the Clustering technique for crime Data Mining.
K-Means Clustering is used to evaluate the performance
of crime scene investigators.
mining technique has been used to discover recurring items in databases in
order to create pattern rules and detect potential future events.
“Association Rule Mining is a method which exploits
the relationships among observations for uncovering crucial information hidden
within big data.”
is proved to be efficient in preventing network intrusions and attacks, such
as DDoS attack.
It also aids in detecting suspicious emails by implementing
Apriori Algorithm. This technique was made to support the investigators in obtaining
information efficiently and hence taking e?ective measures to reduce or curb
This technique is applicable for inspecting
unstructured data to identify common traits among criminal entities.
Classification has been used along with inferential statistics techniques to
predict crime trends. The techniques used are:
It can be used for detecting suspicious emails
and has recorded over 95% preciseness in correctly categorizing emails in a large
data set. The decision tree is produced by the ID3 algorithm.
“The SVM classi?cation is used to detect the
sources of e-mail spamming based on the linguistic patterns of the senders.”
SVM is used for predicting crime hot spots and
analyzing behavior of criminals.
Recently, SVM model has been used to provide
live predictions of crime in urban areas based on Twitter data of the city of
Arti?cial Neural Networks (ANNs) are used to
uncover lies from 371 statements of various crimes. Logistic Regression and
ANNs are used to identify smuggling vessels.